Use of seagrass Zostera novazelandica (Setchell, 1933) as habitat and food by the crab Macrophthalmus hirtipes (Heller, 1862) (Brachyura: Ocypodidae) on rocky intertidal platforms in southern New Zealand

Download Use of seagrass Zostera novazelandica (Setchell, 1933) as habitat and food by the crab Macrophthalmus hirtipes (Heller, 1862) (Brachyura: Ocypodidae) on rocky intertidal platforms in southern New Zealand

Post on 14-Sep-2016




0 download


<ul><li><p>ELSEVIER Journal of Experimental Marine Biology and Ecology, </p><p>214 (1997) 49-65 </p><p>JOURNAL OF EXPERIMENTAL MARINE BIOLOGY AND ECOLOGY </p><p>Use of seagrass Zostera novazelandica (Setchell, 1933) as habitat and food by the crab A4acrophthalmus hirtipes </p><p>(Heller, 1862) (Brachyura: Ocypodidae) on rocky intertidal platforms in southern New Zealand </p><p>Chris M.C. Woods*, David R. Schiel </p><p>Marine Ecology Research Group, Zoology Department, University of Canterbury. Private Bag 4800, Christchurch, New Zealand </p><p>Received 2 February 1996; revised 22 October 1996; accepted 13 November 1996 </p><p>Abstract </p><p>Along rocky reefs at Kaikoura, New Zealand, the crab Macroph~hulrnus lzirripes (Heller, 1862) constructs its burrows exclusively in patches of the intertidal seagrass Zustercc novuzelandiu (Setchell, 1933). The occupation of seagrass patches is widespread, with 63% of patches having crab burrows. The number of externally visible burrows is highly correlated with the number of crabs inhabiting seagrass patches. Plaster casts of burrows revealed that burrows situated at the edge of seagrass patches differed from those away from the edge in having on average more entrances and passageways and, therefore, greater total burrow length. There was an average burrow density of 9.4 (rt 1.04 S.E.) per m2 of seagrass, but burrows were several times as numerous on the edges of patches, especially those that bordered tide pools. The diet of M. hirtipes was found to consist mostly of living seagrass (50.4t3.40% of foregut contents) and sediment (39.2+3.11% of foregut contents), with small amounts of isopods, amphipods and conspecifics also consumed. There were no intersexual or site-specific differences in diet. Ontogenetic differences in diet were found, with smaller crabs consuming more sediment and larger crabs consuming more seagrass. Crabs contributed significantly to erosion of seagrass patches. There was an average of 22.5% (25.27) loss of patch area where crabs were present compared to I .8% (kO.37) where crabs were absent from patches. Laboratory tests showed that burrow construction took around 250% longer in seagrass patches that had denser blade coverage. In the field, burrows persisted for up to 3 months, even if vacated by crabs. Their abundance at patch edges, combined with the loss of sediment and reduced binding of seagrass roots, accelerated erosion of seagrass patches. 0 1997 Elsevier Science B.V. </p><p>*Corresponding author. Present address: National Institute of Water and Atmospheric Research, P.O. Box 14-901, Wellington, New Zealand. Tel.: ( + 64-4) 386.0300; fax: ( + 64-4) 388-9931; e-mail: </p><p>0022.0981/97/$17.00 0 1997 Elsevier Science B.V. All rights reserved P/f SOO22-098 I (96)02767-O </p></li><li><p>50 C.M.C. Woods, D.R. Schiel / J. Exp. Mar. Biol. Ecol. 214 (1997) 49-65 </p><p>Keywords: Burrows; Crab; Macrophthalmus hirtipes; Seagrass; Zostera novazelandica </p><p>1. Introduction </p><p>Seagrasses are marine angiosperms that have successfully adapted to shallow coastal and estuarine environments, fulfilling several important ecological roles (Dawes, 1981; Duarte and Sand-Jensen, 1990). Because of their structural complexity, biodiversity and productivity, they have been described as the marine analog of tropical rainforests (Simenstad, 1994 in Dawes et al., 1995). Seagrasses form extensive continuous meadows, or individually discrete patches (Virnstein, 1995a), providing both a habitat and a source of food for a wide range of organisms including fishes, gastropods, crustaceans, and echinoderms (Kikuchi, 1974; Dawes, 198 1; King, 198 1; Den Hartog, 1987; Perez and Bellwood, 1988; Edgar, 1990; Wassenberg, 1990; Bishop, 1992; Rosas et al., 1994; Dawes et al., 1995; Greenway, 1995; Virnstein, 1995b). Seagrasses may also stabilise bottom sediments, improve water clarity (Ginsberg and Lowenstam, 1958; Dawes, 198 1 ), aid in primary production (Dawes, 1981) and act as substrata for epiphytes (Humm, 1964; Philippart, 1995a,b). </p><p>In New Zealand, seagrasses are represented by the two species Zosteru capricorni and Z. novazelandica (muelleri). Both species are usually found above or at the low tide mark of sheltered muddy tidal estuaries, river mouths and bay heads (Webb et al., 1990). Zostera cupricorni is found only around the North Island of New Zealand, while Z. novazelandica is more widespread, occurring in all regions of the mainland (Webb et al., 1990). </p><p>Around the rocky exposed coastline of the Kaikoura Peninsula, Z. novazelandica occurs in isolated areas upon intertidal siltstone platforms. The seagrass patches initially form in sediment-laden tidal cracks, before expanding along cracks and over the siltstone to form discrete mats (Ramage, 1995). Although these seagrass patches range up to 10 m in size, 75% of them are &lt; 1 m2 (Ramage, 1995), and do not form the extensive beds often seen in more sheltered habitats such as tidal estuaries. The intertidal platforms upon which Z. novazelandica occurs are frequently exposed to high energy oceanic swells and storm waves (Rasmussen, 1965), thus exposing the seagrass patches to physical disturbance. </p><p>The mud crab Mucrophthalmus hirtipes (Brachyura: Ocypodidae) occurs within the patches of Z. novazelundica on the Kaikoura Peninsula. This crab is endemic to New Zealand, most commonly found below the mid-tide level of mud flats of harbours, lagoons and estuaries inhabiting temporary, water-logged burrows. It is in these habitats that such aspects of the biology of M. hirtipes such as burrowing behaviour, feeding, habitat preferences, population structure and reproduction have been investigated (Nye, 1974; Fielder and Jones, 1978; Simons, 1981; Simons and Jones, 1981; Hawkins and Jones, 1982; Jones and Simons, 1982; McLay, 1988). However, it is only through the presence of Z. novazelandica that M. hirtipes can populate the Kaikoura Peninsula, as seagrass patches are the only source of permanent sediment on the intertidal platforms. </p><p>Due to the isolated nature of individual seagrass patches, the habitat for M. hirtipes is </p></li><li><p>C.M.C. Woods, D.R. Schiel I J. Exp. Mar. Biol. Ecol. 214 (1997) 49-65 51 </p><p>more restricted and marginal than that of mudflats in which it usually occurs. Given that previous studies of M. hirtipes occurring on mudflats have revealed it to be a deposit feeder (Beer, 1959; Fielder and Jones, 1978), the restriction of crabs to seagrass patches could affect the diet of M. hirtipes. In this study, we examine the previously uninvestigated use of Z. novazelandica as a resource by M. hirtipes, and the conse- quences of this to seagrass patches on rocky intertidal platforms. </p><p>2. Methods </p><p>2.1. Study sites </p><p>The study was done from May 1994 to July 1995 on the Kaikoura Peninsula on the central east coast of the South Island of New Zealand (4325S, 17342E). This area is at the northernmost position of the Subtropical Convergence (Heath, 1985) and is frequently exposed to high energy oceanic swells and storm waves. The range of annual sea temperatures is 8.5C to 19C (Ottaway, 1976). Two study sites, approximately 3 km apart on opposite sides of the peninsula, were used (Fig. 1). These two sites are the only locations on the Kaikoura Peninsula possessing significant concentrations of seagrass. Wairepo Flats has a northeasterly aspect and is composed of an extensive siltstone platform which gradually slopes down to a subtidal fringe of large brown algae such as Cystophora sp., Corphophyllum maschaloccupum and Macrocystis pyrifera. Intertidally, extensive mats of turfing coralline algae and the fucalean alga Hormosira banskii are interspersed with seagrass patches and bare siltstone patches dominated by limpets and trochid gastropods. Mudstone Bay has a southerly aspect and is similarly composed of an extensive, gradually sloping siltstone platform. The algal and animal assemblages are similar to those at Wairepo Flats. More silt collects at Mudstone Bay, however, due to the shallower sublittoral bed nearby (Rasmussen, 1965). </p><p>Ii5 ii00 IkE ii00 .35 </p><p>40S </p><p>.45 </p><p>B P 8 P </p><p>Al hiudstone Bay Flats </p><p>Fig. I. Map of New Zealand showing the location of the Kaikoura Peninsula and the two study sites. </p></li><li><p>52 C.M.C. Woods, D.R. Schirl I J. Exp. Mar. Biol. Ecol. 214 (1997) 49-6.5 </p><p>2.2. Sampling methods </p><p>Because of the sensitivity of seagrass to disturbance, and the apparent state of decline observed in a number of patches around Kaikoura (Ramage, 199.5) a non-destructive method of quantifying use of seagrass as a habitat by M. hirtipes was necessary. Warren ( 1990) found that for another ocypodid crab, Heloecius cordiformis, inhabiting mangrove forests, counts of open burrows provided a quick and reliable estimate of apparent crab abundance. In May 1994, to determine whether the number of burrows visible on the surface of a patch of seagrass was a reliable indicator of the actual number of crabs inhabiting the seagrass, ten seagrass patches at each site were chosen at random for sampling. A 0.25 m* quadrat was randomly placed within each patch and the number of visible burrows was counted. The area within the quadrat was then rapidly dug out down to the bedrock and placed in a high-walled sieve (diameter of sieve = 60 cm, mesh size = 2 mm). The sediment was washed away from the seagrass and the number of crabs remaining was counted. The crabs collected were sexed to obtain an estimate of the sex ratio at each site. For ten of these 0.25 m2 quadrats (five at each site), the vertical depth of a single burrow selected at random was determined by careful dissection of the seagrass before sieving, measured with a 1 m stainless steel rule and compared with the vertical depth of the seagrass itself. </p><p>In June 1994, transects were run perpendicular to the shore at both sites, from the mean high tide mark to mean low tide, at horizontal intervals of 4 m, along the entire area of each study site (Wairepo Flats = 900 m wide, Mudstone Bay = 450 m wide). The number of seagrass patches encountered 2 m either side of the transect and the presence or absence of burrows in these patches were recorded. To obtain an estimate of overall burrow density, 30 seagrass patches at each site were selected at random and a 1 m quadrat placed at random within the patch. The number of burrows within each quadrat was then recorded. </p><p>The spatial distribution of burrows within seagrass patches was determined by randomly selecting ten large-sized ( &gt; 2 m*) patches at each site. For each patch, three 0.0625 m quadrats were randomly placed and burrows were counted within each of four zones on a patch: O-30 cm from the patch edge, 30-60 cm from the patch edge, 60-90 cm from the patch edge, and 90- 120 cm from the patch edge. As large seagrass patches usually border tide pools along some of their perimeter, the sampling procedure was carried out on both tide pool, and non-tide pool regions of the patches. The percentage cover of seagrass blades was estimated from a grid placed in each quadrat. </p><p>To examine the morphological characteristics of the burrows constructed by M. hirtipes, casts were made of 30 burrows from different patches, 15 at each site. Patches were selected at random, then a random allocation was made as to whether a burrow at the edge (O-30 cm from edge) or a non-edge ( &gt; 30 cm from edge) burrow would be cast. The casts were made when the patches were exposed at low tide using FFFF casting grade Plaster of Paris. Prior to casting, water within the burrows was extracted using a 50 ml syringe with flexible tubing attached which was inserted as far as it would go into the burrow. A liquid slurry of plaster was then poured into the burrows, and the burrows marked with a white upright plastic marker pushed into the seagrass and left for 24 h before the hardened caste was carefully dug out and extracted from the seagrass. Once extracted from the seagrass, the following measurements and observations were </p></li><li><p>C.M.C. Woods, D.R. Schiel I J. Exp. Mar. Biol. Ecol. 214 (1997) 49-G 53 </p><p>recorded for each burrow: greatest vertical depth, total burrow length (length of all entrances and passageways combined), number of crabs per burrow as well as each crabs sex and size, average burrow diameter (mean taken from diameter measured at burrow entrance, mid-point and end-point), number of entrances, and finally, number of passageways per burrow. </p><p>The persistence of crab burrows was determined by placing a 0.25 rn quadrat within each of ten randomly selected patches at each site. The quadrats were marked with plastic corner stakes. A gridded quadrat (100 X 5 cm squares) was laid within the marked area and the position of all existing burrows was mapped. Every 7 d for a further 1 I weeks, the marked areas were examined and all burrows were mapped. </p><p>In August 1994, the effect of blade density on burrowing was determined by selecting I.5 sections (approximately 35 cm X 18 cm X 10 cm deep) which were carefully dug out intact from large seagrass patches ( &gt; I m). Five sections had approximately 100% seagrass blade cover, five had 50% blade cover, and five had no blade cover. Once dug out, these sections were immediately transferred to the laboratory and placed in aquaria (35 cm X I8 cm X I8 cm), covered with 5 cm of fresh seawater, and provided with a bubbled air supply. One crab was placed into each aquarium and the time from initiation of burrow construction to time of burrow completion was recorded. </p><p>To determine the effect of crabs on the rate of patch erosion, each of five patches possessing crab burrows at each site had a white upright plastic marker placed in the centre in December 1994. Five compass bearings were randomly selected, and the distance from the plastic marker to the edge of the patch along each of these compass bearings was measured. The total number of crab burrows present in the patch was recorded, and blade density of the patch was estimated using a 0.25 m gridded quadrat. After 6 months, the distance along each compass bearing from the plastic marker in centre of each patch to the edge of the patch was measured, and number of burrows and blade density were recorded. This treatment included a range of burrow densities from l-48 crab burrows per patch. For comparison, tive patches that did not possess crab burrows at each site were treated as above. </p><p>Sixty crabs were collected from each site for gut content analysis from May to June 1994. Upon capture, crabs were immediately placed in 10% formalin and taken to the laboratory where they were sexed and the carapace width was measured using vernier callipers (20. I mm). The foregut of each crab was dissected out, opened along the mid-ventral line, and the contents washed out into a gridded (100 X 4 mm squares) petri dish. Foregut contents were estimated as the percent volume each different type of food item contributed to the total volume of the foregut contents as estimated by the total number of 4 mm squares covered by the foregut contents. </p><p>3. Results </p><p>3.1. Occupation qfsseagrass patches by crabs </p><p>The numbe...</p></li></ul>


View more >